Design D was Narbeth’s hexagonal-battery ship on roughly a Lord Nelson hull, with reciprocating machinery. It was dated 14 December 1904. For a time it seemed that this design would be chosen, so early in January Narbeth produced a variety of alternative versions.15
Design E (designated G for the Committee on Designs, see below) was yet another twelve-gun ship, arranged on what Narbeth called the Triangle Plan. Guns were mounted in and atop two triangular redoubts pointing towards amidships. A single turret was at the apex of each triangle, superfiring over two pairs of guns firing from each of the two other points of the triangle. This required a ship about as large as C. The main advantage of this configuration over Narbeth’s hexagonal ship was much greater end-on fire against a target crossing the ship’s bow, as all three pairs of 12in guns at either end of the ship could keep firing under those circumstances. This design was never fully examined by the Committee on Designs because broadside fire was far more important than end-on fire. Design F was a cut-down version of the twelve-gun ship, with the highest forward turret and the quarterdeck omitted. That dramatically reduced topweight, so dimensions were reduced to 530ft × 82ft × 26ft (19,000 tons). Compared to the twelve-gun ‘castle’ ship, F cost £1,700,000 rather than £1,940,000. G was the least expensive all-big-gun ship, essentially half of C, with only one ‘castle’, aft. This dramatic reduction cut her to 14,000 tons (425ft × 77ft × 25ft, £1,200,000), but since G cost more than half as much as a full ‘castle ship’, her cost per gun was the highest of the lot. This was the only one of Narbeth’s designs never shown to the Committee on Designs. The lowest cost per gun was D, the Lord Nelson derivative with reciprocating engines (£1,700,000; £142,000 per gun).
On 22 December 1904 Fisher invited a blue-ribbon panel to form a Committee on Designs, nominally to review the types of warships the Admiralty was then proposing.16 He was well aware that his new types of capital ships and destroyers would excite intense criticism and the committee was a way of solving that problem. Fisher deliberately mixed officers who would handle the projected ships with civilian technical experts who could evaluate design issues such as configuration.
Fisher was clearly already determined to buy a ship with an all-big-gun battery and turbine power – turbines might, incidentally, be its most controversial feature. The Committee was asked to consider five types of ships, of which the battleship and the armoured cruiser figure in this book. The only features given were speed (21 knots for the battleship, 25 knots for the cruiser), an all-big-gun 12in battery (nothing between 12in and anti-torpedo guns) and the general standard of armour (‘adequate’ for the battleship, on the scale of the Minotaur for the cruiser). The battleship in particular had to be able to dock at Portsmouth, Devonport, Malta and Gibraltar ‘but the design will not be condemned for the sole reason that the ship cannot be docked at Chatham or pass through the lock there’.
Fisher personally opened proceedings on 3 January 1905. His reported remarks were much those he later used to justify the new type of battleship. He stated that the two governing factors in a battleship were guns and speed. All armament had to be above the upper deck (i.e., not on the ships’ sides) so that it could be fought in any weather and also so as not to interfere with net defence against torpedoes. Existing battleships had central magazines connected to turrets or other guns by horizontal passages. Fisher would reduce vulnerability to underwater attack by providing each turret with its own magazine, at a safe distance from the side of the ship (hence from any mine explosion). Bulkhead penetrations would be eliminated to improve watertight integrity, particularly against the mines which were proving so effective in the Russo-Japanese War. Since Russia and Japan had chosen 20 knots as their future standard, the Royal Navy must choose 21 knots – not as a paper speed, but as an actual speed (it might be necessary to design for 21.5 knots to be sure of making 21 knots).
The Committee was given Designs D, E and F. D was Narbeth’s hexagonal-battery ship. E was Wilson’s vertical-echelon ship (C above) with its quarterdeck omitted. F was a modified version of E with one fewer turret forward. All had Lord Nelson protection. Considerations of blast simplified the choice. Although blast had been only a limited problem in earlier battleships, whose 6in batteries were protected from 12in blast by decks and bulkheads, it now seemed that the blast from one turret could disable gunners in the sighting hood of a neighbouring one. Based on experience in several battleships, the Committee concluded that the sighting hood of a turret should be at least 63ft from the gun muzzles of the next. The naval members decided that, taking into account the length of the 12in gun and the diameter of the turret, turrets should be about 70ft apart (centre to centre). They also argued against superfiring: a lower turret would be untenable in a chase due to blast from the upper turrets. Lower turrets would be tenable only when firing within 20–30° of the beam. Wilson’s ship would offer nothing in ahead or astern fire and the concentration of turrets at each end of the ship, protected by a single redoubt, offered a very large target, the middle turret of the three being an excellent point of aim. Because Design G suffered from the same problems as the other two, it was never shown to the full Committee.
That left Narbeth’s hexagonal Design D. It carried as many guns as Wilson’s ship on 2000 tons less. A modified D1 design had the foremost turret moved up onto a forecastle, to keep it dry. A further modified D2 had the broadside turrets moved further apart, with a large boiler compartment between them. Committee members were provided with a cardboard section of the blast zone, so that they could work out the arcs of fire of the turrets on wooden models of the D designs. It was soon apparent that the close midships turrets of D1 would have very limited arcs of fire. Substituting a single centreline turret for the two after broadside turrets of the D2 design would give better performance. Blast from the two remaining broadside turrets would pass clear of the two centreline turrets aft.
Blast from the two wing turrets would make it impossible to fire the forward centreline turret in a chase. On the available length, it was impossible to move the wing turrets far enough aft for their blast to clear the centreline turret forward. DNC was asked for a further design H, with the five turrets (a centreline turret replacing the two after broadside turrets and the foremost turret on a forecastle). There was no question of replacing the remaining two wing turrets with another centreline turret, because that would have taken up too much centreline space.
Watts preferred D1 because it gave the smallest possible ship with a convenient arrangement of main engines and boilers and the simplest boat stowage. H would require about the same total length of engines, boilers and magazines, but weights were moved more towards the ends of the ship, deck space would be more broken up and boat accommodation more difficult. The Committee briefly favoured D2, but it died because the wide separation of the wing turrets was impractical from a weight balance point of view. The naval members of the Committee, however, unanimously preferred H: efficient gunfire trumped any other consideration. From a fighting point of view, the only sacrifice (compared to D2) was astern fire of one rather than two turrets.
Both Watts and E-in-C urged adoption of turbines, both because of their simplicity and because they would save about 1000 tons. Naval members of the committee wanted fuller information before they agreed. Charles Parsons, who had pioneered steam turbine propulsion, was invited to attend the 17 January meeting of the committee. The main issue was manoeuvrability. Parsons convinced the Committee that he could provide enough reversing power by installing reversing stages on every shaft. The Committee also favoured a suggestion by Froude (of the model basin) that for better manoeuvrability the new battleship should have a fifth propeller shaft, in line with her rudder. This question explains why HMS Dreadnought introduced twin rudders, each in the flow of a propeller, into Royal Navy practice. This practice continued up to the Queen Elizabeth class. Ultimately it was abandoned because it cost speed (the rudders added resistance) and vibration (the rudders vibrated in the flow of the propellers). The issue was revived during the Second World War both because several ships suffered hits aft and because US ships with twin rudders were noticeably more manoeuvrable than their British counterparts (albeit also more subject to vibration).
The Fire Control Committee argued for control positions both above and below the forefunnel. Existing pole masts were not rigid